Sept. 13 (UPI) — Researchers have developed new sequencing techniques and machine learning to more quickly diagnose diseases in newborns, as well as cut down false-positives.
By using their new data analysis method, researchers at Yale were able to examine an entire metabolic profile rather than previous methods that focus on a fraction of data collected.
The research, which was published Wednesday in the journal Genetics in Medicine, found the sequencing method correctly identified 89 percent of newborns with methylmalonic acidemia, an inborn metabolic disorder that can lead to fatal neonatal disease.
In a database review of California newborns, more than 500 babies were falsely identified as having the disorder, compared to 100 that were correctly diagnosed with the condition.
“By combining sequencing with our novel machine learning, we’ve made a big difference in reducing MMA false-positive results,” senior author Curt Scharfe, an associate professor of genetics at Yale, said in a press release.
Families then go through the anxiety of going through a battery of tests to confirm the diagnosis.
“The time until confirmation is stressful for families, places a burden on the health care system and in some cases could delay the right treatment for these infants.” Scharfe said.
Currently, blood drawn from a pinprick of an infant’s heel shortly after birth is analyzed for preventable diseases.
The researchers sequenced 72 metabolic genes from newborn dried blood spots, and analytical and clinical performance was evaluated in 60 screen-positive newborns for methylmalonic acidemia. The researchers also trained a Random Forest machine learning classifier to help improve diagnosis.
The researchers believe the new sequencing and data analysis can be used to complement existing routine blood work to avoid lengthy testing and speed up the treatment for babies.
Using the sequencing and data analysis, the researchers also clinically validated a test for cystic fibrosis.
Ultimately, they said they hope to implement the new testing methods at hospitals worldwide.